And Being implemented it without informing the airlines or the pilots, because it would have required expensive flight simulator sessions for every pilot.
It only used one sensor instead of the standard redundancy of two. When the sensor failed the crew had ten seconds to turn the system off before the plane would enter an unrecoverable dive.
almost.
The extra redundant sensor was an option that the airlines in this case declined.
Any single point of failure that can take down an aircraft is not engineering, it is beyond stupid Bean Counting, but that was the profit driven culture at the time.
When I was building large Datacenters, I spent lots of time looking for these potential single points of failure.
one of my friends used to have nightmares about a bad actor inserting something into “windows update” another was the single security guy for .Net his hair turned grey prematurely.
It only used one sensor instead of the standard redundancy of two.
************
Yep, this what my son the aerospace engineer told me. He works on rockets though, not planes.